Joseph Steven Konrad MD, Presenter: Nothing to Disclose

Derek Merck, Abstract Co-Author: Nothing to Disclose

David Thomas Glidden BS, Abstract Co-Author: Nothing to Disclose

Grayson L. Baird MS, Abstract Co-Author: Nothing to Disclose

Ana P. Lourenco MD, Abstract Co-Author: Nothing to Disclose

Michael David Beland MD, Abstract Co-Author: Consultant, Bard Access Systems

To determine if a textural analysis metric can be implemented to improve diagnosis of adenomyosis by ultrasound.

We retrospectively identified 38 patients with a MRI diagnosis of uterine adenomyosis that also had a pelvic ultrasound within 6 months. We also identified 50 normal pelvic ultrasound exams confirmed by a normal pelvic MRI within 6 months as a control group. A region of interest (ROI) was subsequently placed on the study population ultrasound image corresponding to the area of adenomyosis on MRI. A ROI was placed in the area of the junctional zone in the normal controls. The abnormal and normal ROIs were then filtered to produce several metrics of texture variability and compared against trained normal and abnormal distributions to determine the success rate, sensitivity, specificity, negative and positive predictive values. The ultrasound reports performed prior to MRI were also reviewed to determine the radiologist false negative rate for comparison to our textural analysis metric.

Using a training population of 50 normal ultrasound exams (confirmed with a normal MRI) and 38 abnormal ultrasound exams (MRI confirmed adenomyosis) we had an overall 75% (66/88 accurately diagnosed) success rate with a sensitivity, specificity, negative and positive predictive values of 70%, 79%, 73%, 76%, respectively (p<.0001). The false negative rate of the initial ultrasound interpretation was 74% (28/38). 

An easily applied uterine textural analysis of pelvic ultrasound images can accurately diagnose adenomyosis.

Further development in textural analysis may allow radiologists to make a definitive diagnosis of adenomyosis with ultrasound, precluding the need for a confirmatory MRI.

Konrad, J, Merck, D, Glidden, D, Baird, G, Lourenco, A, Beland, M, Improving Ultrasound Detection of Uterine Adenomyosis through Computational Texture Analysis .  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14019591.html